ultrathink
Pass
Audited by Gen Agent Trust Hub on Feb 22, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Prompt Injection (SAFE): No instructions to bypass safety filters or ignore system rules were detected. The 'Reality Distortion Field' and 'Think Different' sections are stylistic persona markers rather than malicious overrides.- External Downloads & RCE (SAFE): The skill contains no package installations (npm/pip) or remote script execution patterns.- Data Exfiltration (SAFE): No network operations, hardcoded credentials, or sensitive file path access were identified.- Indirect Prompt Injection (LOW): The skill encourages the agent to read Git history and external codebases. This creates an ingestion surface for untrusted data. Evidence: 1. Ingestion points: Git history, existing codebase; 2. Boundary markers: Absent; 3. Capability inventory: General mention of bash tools and MCP servers; 4. Sanitization: Absent.
Audit Metadata